Skip Navigation

Annals of Botany 2007 99(1):171-182; doi:10.1093/aob/mcl240
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Greenwood, D. J.
Right arrow Articles by Mulholland, B. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Greenwood, D. J.
Right arrow Articles by Mulholland, B. J.
Agricola
Right arrow Articles by Greenwood, D. J.
Right arrow Articles by Mulholland, B. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?


© The Author 2006. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

A Novel Method of Supplying Nutrients Permits Predictable Shoot Growth and Root : Shoot Ratios of Pre-transplant Bedding Plants

Duncan J. Greenwood*, John M. T. Mckee, Deborah P. Fuller, Ian G. Burns and Barry J. Mulholland{dagger}

Warwick HRI, Warwick University, Wellesbourne, Warwick CV35 9EF, UK

* For correspondence. E-mail d.greenwood{at}warwick.ac.uk

Received: 18 August 2006    Returned for revision: 18 September 2006    Accepted: 27 September 2006   


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 APPENDIX 1
 APPENDIX 2
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 

BACKGROUND AND AIMS: Growth of bedding plants, in small peat plugs, relies on nutrients in the irrigation solution. The object of the study was to find a way of modifying the nutrient supply so that good-quality seedlings can be grown rapidly and yet have the high root : shoot ratios essential for efficient transplanting.

METHODS: A new procedure was devised in which the concentrations of nutrients in the irrigation solution were modified during growth according to changing plant demand, instead of maintaining the same concentrations throughout growth. The new procedure depends on published algorithms for the dependence of growth rate and optimal plant nutrient concentrations on shoot dry weight Ws (g m–2), and on measuring evapotranspiration rates and shoot dry weights at weekly intervals. Pansy, Viola tricola ‘Universal plus yellow’ and petunia, Petunia hybrida ‘Multiflora light salmon vein’ were grown in four independent experiments with the expected optimum nutrient concentration and fractions of the optimum. Root and shoot weights were measured during growth.

KEY RESULTS: For each level of nutrient supply Ws increased with time (t) in days, according to the equation {Delta}Ws/{Delta}t=K2Ws/(100+Ws) in which the growth rate coefficient (K2) remained approximately constant throughout growth. The value of K2 for the optimum treatment was defined by incoming radiation and temperature. The value of K2 for each sub-optimum treatment relative to that for the optimum treatment was logarithmically related to the sub-optimal nutrient supply. Provided the aerial environment was optimal, Rsb/Ro{approx}Wo/Wsb where R is the root : shoot ratio, W is the shoot dry weight, and sb and o indicate sub-optimum and optimum nutrient supplies, respectively. Sub-optimal nutrient concentrations also depressed shoot growth without appreciably affecting root growth when the aerial environment was non-limiting.

CONCLUSION: The new procedure can predict the effects of nutrient supply, incoming radiation and temperature on the time course of shoot growth and the root : shoot ratio for a range of growing conditions.

Key words: Root: shoot ratio, growth rate, nutrients, nitrogen, phosphate, potassium, evapotranspiration


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 APPENDIX 1
 APPENDIX 2
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
Bedding plants for the retail market are produced by specialist nurseries as seedlings in small ‘plugs’ of peat-based media (plug plants). These small plugs retain only limited amounts of nutrients so that seedling growth is heavily dependent on the nutrients supplied in irrigation solution. Such seedlings need to grow quickly (to maximize productivity), show no nutrient deficiency symptoms, and have good root systems so that after 25–40 d they can be efficiently transplanted into larger containers by robotic transplanters. High root : shoot ratios are especially important to ensure cohesion of the plugs during transplanting, and to ensure good take-off after transplanting (Zandstra and Liptay, 1999). In commercial systems, shoot growth is suppressed by applying plant growth regulators in order to allow time for the root system to develop to an adequate size for successful transplanting. There are, however, environmental and cost pressures against the use of plant hormones, and a promising alternative approach appears to be to modify the nutrient supply.

In most experiments on this topic a range of different concentrations of a standard nutrient mix are applied throughout growth, despite changing plant needs. Responses are considerable but appear to differ with the time of year (Kang and van Iersel, 2001). There is little evidence about their reproducibility so their value for giving practical advice is limited. Salinity can increase with time when using fixed concentrations (van Iersel, 1999; James and van Iersel, 2001) which may depress growth and occasionally induce quality defects. Attempts to overcome these problems have included the use of slow release fertilizers (van Iersel et al., 1999; Stamps, 2000), and monitoring nutrient concentrations in plant tissue (van Iersel et al., 1999; Stamps, 2000) and plug extracts (Scoggins et al., 2002). An alternative approach is to develop a system in which the nutrient supply could be adjusted throughout growth to meet the changing plant demands for nutrients.

Several requirements need to be met to achieve this. First, such a system must take account of the fact that seedlings are grown close together, so that their specific growth rate falls towards the end of the growth period, even under well-nourished conditions. A simple equation is needed that allows the time course of dry matter increase to be estimated over both the exponential and linear phases of growth. One such equation defines the dependence of growth rate on plant mass per unit area in terms of a single growth rate coefficient. It was found that the value of this coefficient remains almost constant throughout the growth of many UK field vegetable crops (Greenwood et al., 1977). Such an equation appears to offer a concise means of summarizing the time course of increase in dry weight of bedding plants receiving optimum nutrient supplies, especially as it might be adjusted for differences in radiation and temperature (e.g. Scaife et al., 1987; Brewster and Sutherland, 1993). A second requirement is to find a means of estimating the decline, with increase in plant mass, in the optimum plant nutrient concentration which is defined as the lowest concentration compatible with maximum growth. A number of equations have been used to describe these relationships (e.g. Lemaire, 1997; Greenwood and Stone, 1998; Knecht and Göransson, 2004). Then, by using an approach that combines these equations, it should be possible to estimate plant mass and the optimum plant nutrient concentrations simultaneously throughout growth and, in consequence, to estimate how to adjust nutrient supplies so as to meet changing plant nutrient demands as they grow. Moreover, by supplying only a fraction of the optimum nutrient supply, the growth rate coefficient might be reduced to a lower value, with a concomitant increase in the root : shoot ratio (Ericsson, 1995). The appearance of nutrient deficiency symptoms would be unlikely in view of the evidence that these only occur after a sudden decline in plant nutrient concentration (Ingestad and Lund, 1979; Ericsson and Ingestad, 1988). Excellent crop appearance could therefore be sustained.

Exploring these possibilities could lead to improved efficiency in the production of bedding plants. In commercial practice the choice of nutrient supply depends on financial, management and organizational considerations of the individual production unit as well as requirements of the plants. To aid making this choice, there is a need to obtain a general means of forecasting the effects of nutrient supply on the growth rate and the root : shoot ratio that applies in widely different conditions. The simplest solution is to provide graphs showing the expected effects of sub-optimum supplies of nutrients on the time course of the shoot weight and the root : shoot ratio, which can be interpolated as required. The dry weight of seedlings of interest to growers depends on the market requirement but is within the range 80–150 g m–2. There is, however, little interest in the concentrations of plant nutrients unless they affect the appearance of seedlings.

The objectives of the study were: (a) to devise a system of growing bedding plants that enabled the growth rate coefficient to be maintained at a constant value with the optimum nutrient supply and a lower but constant value for a sub-optimum nutrient supply, without the appearance of nutrient deficiency symptoms; (b) to find whether sub-optimal growth rate coefficients resulted in increased root : shoot ratios, and to devise an equation for relating it to plant nutrient stress; and (c) to provide a means of forecasting, from readily available information, the effects of different rates of nutrient supply on the shoot and root : shoot ratios throughout the growth period


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 APPENDIX 1
 APPENDIX 2
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
The new procedure for optimizing nutrient supply to meet changing crop demand depends on the published equations described below. An outline of the actual procedure and how it is adapted to provide sub-optimum levels of nutrients will be given. Details of the experiments employed are described next, followed by equations that are used to interpret the results and, in particular, to elucidate the effects of nutrition on the root : shoot ratio.

Equations to optimize nutrient supply to meet changing crop demand
Definitions of the symbols and the units are given in Appendices 1 and 2.

Shoot dry weights
The rate of shoot dry matter increase {Delta}Ws/{Delta}t was initially exponential and then tended towards linear as these close-spaced seedlings grew. The kinetics of growth, until the onset of senescence, have previously been represented by


Formula 240M1

(1)
where Ws (g m–2) is shoot dry weight, K2 is the growth rate coefficient (g m–2 d–1) and K1 (g m–2) is the value of Ws at which the rate of increase of shoot weight is half the maximum (Greenwood et al., 1977; Smolders and Merckx, 1992; Smolders et al., 1993). The coefficients K1 and K2 (g m–2 d–1) are highly correlated and K1 was always set at 100 g m–2 as this gave good fits to field data (Greenwood et al., 1977). When K1 is set equal to 100 g m –2 and {Delta}t->0 then integration of eqn (1) gives


Formula 240M2

(2)
where Ws=Ws0 at the start, at time t0, and Wse is the shoot dry weight at a later time, te.

Optimum nutrient concentrations in the shoots
These must be corrected for their decline as Ws increases (Lemaire, 1997). Nutrient concentrations in the shoot dry matter were calculated from factorial experiments in which widely different levels of N, P and K were applied and there was a single harvest (not reported). Treatments giving the values of Ws within 5 % of the maximum were identified. The average Ws and %N, %P and %K were determined and assumed to be the optimum %N (Nopt), the optimum %P (Popt) and the optimum %K (Kopt) in the shoot dry matter at this Ws. These values were used to calculate values of the ratios Popt : Nopt and Kopt : Nopt. They were taken to be 0·18 and 1·58 for both pansies and petunias. These values were assumed to remain constant throughout growth in view of previous evidence (Ericsson and Ingestad, 1988; Greenwood and Stone, 1998; Belanger and Richards, 1999; Broadley et al., 2004; Knecht and Göransson, 2004). The values of Nopt and the corresponding values Ws were substituted in eqn (3)


Formula 240M3

(3)
to give Bn(c) where (c) indicates that Bn varies with the species. Bn had a value of 3·05 for pansy and 2·81 for petunia.

The resulting equation is used to calculate values of Nopt for different values of Ws, in the new procedure, with the corresponding values of Popt and Kopt calculated from the Nopt and the above ratios. Equation (3) was obtained by replacing the critical %N with Nopt in a previously published formula (Greenwood and Stone, 1998) for the decline in critical %N with increase in Ws.

Weekly average concentrations of N in the irrigation solution
This is calculated from the expected Ndemand and the expected evapotranspiration. Ndemand is calculated from


Formula 240M4

(4)
where (d) indicates a particular day, the 0·9 is included to correct, approximately, for N in the root and the 100 to correct for Nopt being in percentages. The average of the daily concentrations of N required in the irrigation solution was calculated as


Formula 240M5

(5)
where Evap(av) is the expected average daily evaporation in g m–2 (i.e. mm d–1) for the coming week and the 106 to convert Nconc from g of N g–1 of evapotranspiration to µg of N (g of water)–1 or mg of N L–1. The values of Evap(av) are determined as described in the following sub-section. The average weekly concentrations of P are set equal to 0·18 Nconc and those of K to 1·58 Nconc, equivalent to the ratios of Popt/Nopt and Kopt/Nopt in the plant dry matter, and assume there are no selectivity effects during uptake. The same stock solution was used for both species. It was diluted to meet the predicted Nconc(av) as defined by eqn (5). Details of other nutrients present are given in the description of the individual experiments.

The new procedure
First the procedure for providing the optimum supply of nutrients by the ebb and flood method (described below in detail) to meet the changing needs of seedlings throughout their growth is described. The procedure depends on the finding (Fig. 1) that the amount of irrigation applied is equal to that lost by evapotranspiration and the assumption that the aerial environment for a given week is approximately the same as in the previous week. At the start, seedlings relied on nutrients initially present in the plugs and seeds. After a short period, the optimum nutrient supply was estimated by repeating the following steps, each week, to ensure that growth was never limited by lack of nutrients.

Step 1. Determine the shoot dry weights at the beginning and end of the week and the average daily transpiration rate over the entire week.
Step 2. Calculate the growth rate coefficient K2 (g m–2 d–1) by substituting the shoot dry weights in eqn (2). Calculate the expected daily shoot dry weights for the next week from K2 and eqn (1).
Step 3. Predict the demand of N for each day of the next week from eqn (4) by substituting the expected dry weights (from Step 2) and their optimum shoot N concentrations calculated from their shoot dry weight by eqn (3).
Step 4. Predict the expected average optimum N concentration in the irrigation solution for the next week from eqn (5) by dividing the expected N demand for each day of the week (from Step 3) by the expected average daily evapotranspiration (from Step 1) and taking the average for the week.
Step 5. Calculate the concentrations of P and K in the irrigation solution from the average optimum N concentrations in Step 4 (using the same fixed P : N ratio and the K : N ratios throughout growth), and apply the required irrigation solution each day of the week.
Step 6. Repeat each of the above steps for each week of the growth period.


Figure 2401
View larger version (7K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 1. Relationships between measured amounts of chloride in peat plugs and those calculated as the product of chloride concentration in the irrigation solution and loss of water by evaporation. The regression is y=1·039x, r2=0·991. Measurements were made in an experiment in which un-cropped trays were irrigated with solutions of NaCl (4 and 8 mM) for between 1 and 4 d under conditions that were otherwise identical with those for irrigating with nutrient solution.

 
The calculations are thus dependent on measurements of the shoot dry weight with the optimum treatment.

Sub-optimum nutrient treatments were studied to elucidate their effects on the root : shoot ratio. The procedure for maintaining sub-optimal nutrient concentrations was the same as that described above except that irrigation each week was with a solution containing a fraction of the optimum N concentration (determined as described above) and the same P : N and K : N ratios.

Details of experiments
Seedlings were grown in standard commercially available 360 plug (cell) trays (0·48x0·28 m). Each cell had a volume of about 5 mL, was open at its surface but had a drainage hole at its base. Cells were filled with unfertilized sphagnum peat, limed to pH 5·5 (Bulrush Peat Co Ltd, Bellaghy, Magherafelt, Northern Ireland). This had a cation exchange capacity of about 100 meq L–1, a bulk density of about 100 kg m–3 and initially contained water soluble concentrations that, depending on the experiment, were between 42 and 76 mg L–1 of peat for (NH4+NO3)-N, 18 and 29 mg L–1 of peat for K and always <0·6 mg L–1 of peat for P. A single seed was sown in each cell, the trays were watered, stacked on pallets, wrapped with cling film and incubated in the dark at 16 °C for 5 d. Trays were then transferred to a glasshouse in which there was supplementary lighting with high pressure sodium lamps in the January and March experiments, but not those in April and June. Supplementary lighting was set to start when the light levels dropped below 24 W m–2 (Par) and to switch off when they reached 49 W m–2 (Par) between 0100 and 1300 h. Other environmental conditions are given in Table 1. The trays were on ebb and flood benches (2x0·3 m) with 5-cm sides with four holes at their base connected to tubing through which nutrient solution could be pumped. The trays were irrigated each day by raising the level of irrigation solution until the surface of the compost glistened indicating near saturation. After 1 min, the level of irrigation solution was lowered and the cells drained. For about 10 d the irrigation solution contained no nutrients and seedlings relied on residual nutrients in the peat. Thereafter nutrient concentrations were adjusted each week according to plant demand by diluting a stock solution that contained N : P : K : Ca : Mg in the ratio 100 : 18 : 158 : 23 : 11, where N was mostly NO3-N and the remainder was NH4-N. The same stock solution was used for both species. It was diluted each week to meet the predicted Nconc(av) as defined in eqn (5). Suitable aliquots of a stock solution of minor elements were also added to the irrigation solution to maintain the same concentrations throughout each experiment.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Experimental details

 
Four experiments were carried out with sowing dates in January 2005, March 2004, April 2005 and June 2005 subsequently referred to as Jan05, March04, April05 and June05. Details of the experiments, including glasshouse conditions, are given in Table 1. Two bedding plant species were grown in each experiment: pansy, Viola tricola ‘Universal plus yellow’ and petunia, Petunia hybrida ‘Multiflora light salmon vein’. In each experiment, plants were grown with the optimum nutrient regime (prepared as already described in the sub-section ‘New procedure’). Further treatments providing a fraction (typically 0·25 or 0·125) of the predicted optimum concentrations for N, P and K were given to provide sub-optimal nutrition. A treatment exceeding the optimum was included in the June05 experiment. Shoot and root fresh weights were determined at weekly intervals and dry weights established after drying for 48 h at 80 °C.

Equations used in the interpretation of the results
Effect of temperature and radiation on K2
When K2 (g m–2 d–1) was measured with the optimum levels of nutrients it was considered that it might be related to temperature and incoming radiation by an adaptation of equations developed by Scaife et al. (1987) and Brewster and Sutherland (1993). It is


Formula 240M6

(6)
where Tm is the mean daily temperature, Tz is the base temperature when growth ceases, Par (W m–2) is the photosythetically active radiation, calculated as 0·45xtotal solar radiation measured with a Kipp's solarimeter, f is a coefficient that determines the relative effects of temperature and solar radiation on K2 (g m–2 d–1) and g is a proportionality constant. The parameters f, Tz and g were estimated from the results of a controlled-environment experiment kindly provided by Dr A. Langton of Warwick HRI. The experiment tested the effects of three temperatures, 14 °C, 18 °C and 22 °C, in factorial combination with two photosynthetically active radiation intensities (1·73 and 3·46 MJ m–2) on the shoot dry weights of pansy and petunia at three times during the growth period. The values of the parameters are given in Appendix 1.

Root : shoot ratio
Detailed models have been advanced to improve understanding of the effects of various stresses on the root : shoot ratios (e.g. Brouwer, 1962; Thornley, 1972; Ericsson, 1995; Tinker and Nye, 2000, p. 266; Ågren and Franklin, 2003). Although they require more inputs than can be readily obtained, they provide the basis of a relationship with which to interpret the bedding plant data produced in the present study. It is based on the view that nutrient deficiency of N, or P but not K (Ericsson, 1995) around roots restricts shoot more than root growth and conversely a sub-optimal aerial environment restricts root growth more than shoot growth. Equations describing these relationships are derived below.

It is hypothesized that if R is the ratio of root : shoot dry weight and Sn and Sa are the nutrient and aerial stresses then


Formula 240M7

(7)
where {alpha} and ß are coefficients.

If sub-optimum nutrient supply is denoted as subscript sb and optimum nutrient supply is denoted as subscript o then dividing the above equation for the sub-optimum treatment by the corresponding equation for the optimum treatment gives


Formula 240M8

(8)
For a given aerial environment (Sa)sb/(Sa)o is a constant equal to {gamma} so that eqn (8) can be written as


Formula 240M9

(9)
A measure of the nutrient stress integrated over time is (Ws)o/(Ws)sb which may therefore be set equal to (Sn)sb/(Sn)o. As (Sn)o is by definition equal to 1, it follows from eqn (9) that


Formula 240M10

(10)
where {gamma}=1, if the aerial environment is optimum and <1 when it is sub-optimum.

If Wr is root weight, then substitution of Rsb=[(Wr)sb]/[(Ws)sb] and Ro=[(Wr)o]/[(Ws)o] in eqn (10) gives


Formula 240M11

(11)
Accordingly, if the aerial environment is near optimum, limiting the supply of nutrients would be expected to produce plants that had reduced shoot growth but the same root growth compared with plants receiving the optimum supply of nutrients.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 APPENDIX 1
 APPENDIX 2
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
Dependence of the growth rate coefficient on nutrient concentration
In every experiment, Ws increased in a curvilinear manner with time over the period for which different nutrient regimes were imposed, and 100lnWs+Ws was linearly related to time (Fig. 2). The gradient of the linear relationship is the value of K2 (g m–2 d–1) in eqns (1) and (2), and the corresponding value of r2 is a measure of the extent to which it remains constant over the growing period. The degree of fit between the time of imposing the nutrient regime and the final harvest was always very high, as is indicated by r2{approx}1, see the representative data given in Table 2. The coefficient of variation of K2 (g m–2 d–1) for each treatment was about 4 %. This indicates that K2 (g m–2 d–1) was almost constant for both the optimum and sub-optimum treatments, and thus that eqns (1) and (2) gave good descriptions of the time course of shoot dry weight increase for every data set. The gradients always decreased with decrease in nutrient supply. The ratios of sub-optimum to optimum values of K2 (g m–2 d–1) were related to the corresponding ratios of nutrient supply by a single logarithmic relationship that covered all species and experiments (Fig. 3). Increasing the nutrient supply from the optimum to 1·5x the optimum in the June 05 experiment increased K2 (g m–2 d–1) only slightly from 22·3 to 23 for pansy and from 25·9 to 26·8 for petunia (Table 2). Thus growth of Ws was depressed by restricting nutrient supply below the optimum, but was hardly affected by increasing it. The optimum nutrient regime therefore achieved near maximum growth so the method produced near optimum conditions. Values of K2 (g m–2 d–1) for both optimal and sub-optimal regimes were almost constant for long periods. In addition there were no visual signs of nutrient deficiency.


Figure 2402
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 2. (A) Relationships between shoot dry weight (Ws) and the number of days from sowing for pansies grown in the Mar04 experiment, with 1x, 0·5x, 0·25 and 0·125x optimum supply of nutrients denoted by open circles, closed circles, open squares and closed squares, respectively. The standard errors of the difference of the means of Ws after 23, 30, 37 and 44 d from sowing were 0·63, 5·2, 1·64 and 6·0, respectively, with 3 d.f. in each case. (B) Linear relationships between 100lnWs+Ws and time (d) for the data in (A); the gradients (K2) obtained with 1x, 0·5x, 0·25x and 0·125x the optimum supply were 18·9, 15·8, 12·7 and 9·9 (g m–2 d–1), with the corresponding standard errors 0·49, 0·84, 0·59 and 0·42, respectively.

 


Figure 2403
View larger version (9K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 3. The ratios of sub-optimum to the optimum values of K2 (g m–2 d–1), and the corresponding ratios of nutrient supply for pansy (open circles) and for petunia (closed circles) grown in each experiment. y=0·182lnx+0·9833, r2=0·82.

 


View this table:
[in this window]
[in a new window]

  		TABLE 2. The degree of fit (r2) and the gradient (K2) of the linear regression of 100lnWs+Ws against time for the period during which different nutrient regimes were imposed

 
For the optimum treatment of each species and each experiment 100lnWs+Ws was regressed against cumulative evapotranspiration from emergence to the final harvest. This latter regression gives the growth rate coefficient, K2evap, expressed in terms of daily millimetres of evapotranspiration instead of time, as in K2 (g m–2 d–1). An example of the comparisons between K2 (g m–2 d–1) and K2evap is given for petunia in Table 3. The values of r2 were always between 0·96 and 1·0 but K2 (g m–2 d–1) was generally estimated more accurately than K2evap, with average coefficients of variation of 3·7 % and 7·0 %, respectively. Values of K2 (g m–2 d–1) for the June05 experiment were higher than for the other experiments but K2 (g m–2 d–1) did not vary greatly between the Jan05, March04 and April05 experiments. There was also little variation between values for K2evap for the March04, April05 and June05 experiments which were always about 13·5 g m–2 mm–1. The values were higher for the Jan05 experiments possibly because of the lower evapotranspiration.


View this table:
[in this window]
[in a new window]

  		TABLE 3. The gradients K2 and K2evap and the degree of fit (r2) of the linear regressions of 100lnWs+Ws and against time and against cumulative evapotranspiration, respectively, from emergence to final harvest for petunia

 
Effect of aerial environment on K2
The values of K2 (g m–2 d–1) for the optimum nutrient treatment of each experiment on both pansies and petunia were plotted against the values calculated by eqn (6) from the average temperature and radiation over the growth period (Fig. 4). A near proportional relationship existed between the two sets of values and there was little difference between pansy and petunia.


Figure 2404
View larger version (9K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 4. Relationship between values of K2 (g m–2 d–1) measured with the optimum supply of nutrients and the values calculated from the average values of Par and temperature from emergence to final harvest using eqn (4) calibrated with data from independent controlled environment (CE) experiments. Data for pansy denoted by open circles and for petunia by closed circles; y=1·159x, r2=0·78;. Tz=0 for pansy and +3 °C for petunia.

 
Root : shoot ratio
The validity of the quantitative relationship (eqn 10) for the dependence of the root : shoot ratio on nutrition was supported by the data from all experiments except the Jan05 experiment. A proportional relationship existed between Rsb/Ro and [(Ws)o]/[(Ws)sb] for the Mar04, April05 and June05 experiments on pansies and on petunia (Fig. 5). The relationships were little affected by the experiment or the species as is indicated by similarity in the residual mean squares after fitting (not shown). The same proportional relationship fitted the entire data set and had a gradient 0·91 which suggests that the root : shoot ratios were not appreciably affected by inadequate aerial environments (Fig. 5). As the gradient, and thus {gamma} was almost constant it follows from eqn (10) that nearly all the variation in the root : shoot ratio can be attributed to the effect of nutrient supply on shoot weight. This is illustrated by data from the March04 experiment in Table 4.


Figure 2405
View larger version (9K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 5. Dependence of the root : shoot ratio of pansy and petunia on the reciprocal of the inhibition of growth induced by nutrient stress. The experiments were Mar04 (open circles), April05 (closed circles) and June05 (closed squares). There was no significant difference between the proportional relationships for the different experiments, and the line of best fit for the entire data set for both crops is y=0·912x, r2=0·64 (d.f. 52). It is drawn on both graphs.

 


View this table:
[in this window]
[in a new window]

  		TABLE 4. Effect of sub-optimum nutrient supply on shoot and root growth of petunia in experiment Mar04

 
The petunia data in the Jan05 experiment had a gradient of 0·66 and an r2 of 0·73, but there was no proportional relationship for pansy and only a weak linear one (r2=0·27). For these experiments, {gamma} of eqn (10) was variable and considerably <1. According to the theory, this indicates that the aerial environment limited growth, which may have resulted in part by reliance on supplementary lighting. The latter has a different spectrum of wave-lengths from natural incoming radiation which may have affected the root : shoot ratio (Von Arnim and Deng, 1996).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 APPENDIX 1
 APPENDIX 2
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
The new method of supplying nutrients
The dependence of the growth rate coefficients on nutrient supply, and their near constancy over long periods for a given supply, is crucial to the approach adopted in the present study. To further explore the reasons for this constancy a model was written based on a simplified view of the various processes. It was developed from eqns (1) and (3) with the approximation that for each day the ratio of K2 (g m–2 d–1) for the sub-optimum treatment to that for the optimum treatment is equal to the ratio of shoot %N to the optimum shoot %N (Nopt). The optimum value of K2 (g m–2 d–1) is a constant and the sub-optimum and optimum treatments have the same starting values of Ws and Nopt (calculated from eqn 3). For each day the increment of Ws and of N-uptake for the optimum level of N is calculated from eqn (1) and eqn (3) using the optimum value of K2 (g m–2 d–1); the total weight is obtained by adding the increment to the previous day's total weight and likewise for the uptakes. The sub-optimum uptakes are calculated similarly except that for each day the increment in sub-optimum uptake is a constant fraction of the uptake with optimum treatment. It is added to the total sub-optimum uptake on the previous day, a new value of sub-optimum %N is calculated, and an up-dated sub-optimum value of K2 (g m–2 d–1) is calculated from the ratio of the sub-optimum %N/optimum %N and the optimum value of K2 (g m–2 d–1); the increment in sub-optimum Ws is then calculated from eqn (1). It is assumed that P and K never limited growth. A summary of significant simulations are given in Fig. 6. Figure 6A gives the time course Ws for optimum and sub-optimum (0·375x optimum) supply of N; regression of 100lnWs+Ws against time in each case had an r2>0·999 which illustrated that K2 (g m–2 d–1) was almost constant. When supply was sub-optimum there was a short initial period of instability as the %N in the plant adjusted from a high value to a lower one. But afterwards, K2 (g m–2 d–1) and %N were almost constant despite a large increase in Ws (Fig. 6B), although there was a slight decline in %N with increase in Ws that mirrored the decline in optimum %N (Fig. 6B). It therefore appears that the near constancy of K2 (g m–2 d–1) at both optimum and sub-optimum levels can be largely explained in terms of published algorithms.


Figure 2406
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 6. (A) The simulated dependence of Ws on time for seedlings that received the optimum or sub-optimum (0·375x optimum) daily supply of N in the irrigation solution. (B) The corresponding simulated relationships between values of %N in the shoot dry matter and the sub-optimum values of K2 (g m–2 d–1) on Ws. The crop parameters were Bn=2·93 and K2=20·54 g m–2 d–1, which are the averages for pansy and petunia. All simulations started with Ws=1 g m–2 d–1 and %N=5·3.

 
The assumption in the method that the variation in the environment from one week to the next would not cause significant error in the estimation of nutrient concentrations in the irrigation solution is supported by the relatively small inter-experiment variation in K2 (g m–2 d–1) when nutrient supply was optimum (Table 2). The present experiments (e.g. Fig. 1) indicate that the increase in nutrients retained in the plugs available for plant uptake is the product of evapotranspiration and concentration of nutrients in the irrigation solution; a finding that was confirmed for a commercial overhead irrigation system where there was little run-off (D. J. Greenwood et al., unpubl. res.). This implies that the method described in this paper could apply to this system as well as the ebb and flood experimental system.

Evidence on the extent to which nutrient uptakes in the shoots plus roots in the optimum treatments matched the supply of nutrients in the irrigation solution was obtained from two experiments. For each N, P and K nutrient, the total uptake in the shoots plus roots was expressed as a fraction of the cumulative amount of that nutrient supplied by irrigation. If there was perfect agreement the fraction would be one. The values of the fraction in each experiment were averaged over the last two harvests of pansies and petunias. The average values of the fractions for N, P and K were 0·80, 1·07 and 0·99 for the Mar04 experiment and the fractions were 0·94, 1·16 and 0·83, respectively, for the June05 experiment. It is possible that the ratios were influenced by chemical and biological processes in the compost and by harvest not being synchronized with the weekly changes in nutrient concentrations in the irrigation solution. Even so the evidence supports the view that there were no serious imbalances between supply and plant demand of the major nutrients.

Evidence that the new procedure provides near optimum nutrient conditions and thus gives near maximum growth rates is provided by the fact that values of K2 (g m–2 d–1) were reduced by lowering nutrient concentrations below the optima but only slightly increased by increasing them above it (Table 2). Additional evidence is provided by the values of Bn being well within the range of values expected from the literature (Appendix 1 and Table 5). The ratios of P : N and K : N used for estimating P and K concentrations in the irrigation solution are rather higher than the published mean values for other species (Appendix 1 and Table 5), which suggests that there is luxury consumption of these nutrients, and that the concentrations of P and K could be reduced without any restriction in growth, although some of the discrepancies might be associated with the published ratios having been derived for larger plants. The insensitivity of bedding plant growth to substantially increasing all nutrients above their optima (Table 2) also implies that the validity of the method is little affected by some accumulation of excess nutrients in the compost. The average value of K2 (g m–2 d–1) with the optimum nutrient supply (Table 2), though varying with the aerial environment is also within the range of those found for field vegetables grown, with adequate nutrients and water, in the UK between April and September (Table 5). The values of K2evap, calculated by replacing time with cumulative evapotranspiration are, however, only about half those for field vegetables (Greenwood et al., 1977) which indicates a much greater water use efficiency in the glasshouse than in the field, as might be expected from the higher humidities and lower wind speeds.


View this table:
[in this window]
[in a new window]

  		TABLE 5. Independently measured parameter values

 
The new method of supplying nutrients allowed K2 (g m–2 d–1) to be maintained at a given sub-optimum value for a long period (Fig. 2 and Table 2) without the appearance of any deficiency symptoms. This supports the view that, over these periods, the system did not lead to appreciable fluctuations in nutrient supply as such symptoms generally only occur when the concentration of a nutrient in the plant declines suddenly (Ingestad and Lund, 1979). This could be important for the horticulture industry because it may enable plants to be grown with high root : shoot ratios without showing deficiency symptoms.

Relationship between root : shoot ratio, nutrient supply and time
The present approach to providing predictive graphs was based on the constancy, over long periods, of nutrient-supply dependant values of K2 (g m–2 d–1), as this enabled the shoot dry weight for each day to be calculated for each nutrient regime. A simulation model to make such calculations was programmed on the basis of calibrated versions of eqns (1), (6) and (10). The increment in dry weight was calculated for each day by eqn (1). The value of K2 (g m–2 d–1) in this equation was calculated from the mean temperature and Par by eqn (6), with the values of f and g given in Appendix 1. These values of K2 (g m–2 d–1) for optimum nutrient supplies were corrected to give the value for any sub-optimum nutrient supply by the relationship given in Fig. 3. The value of Rsb/Ro was calculated from the ratio of Ws for the sub-optimum nutrient supply to that for the optimum supply by the proportional relationship obtained in Fig. 5. They show (Fig. 7A) that the root : shoot ratio was only appreciably increased when the nutrient supply was <0·6x the optimum, and then it increased in a diminishing manner to a maximum with increase in time. With a nutrient supply of 0·4x the optimum, the root : shoot ratio increased by a factor of 1·4 after 25 d when Ws was equal to 100 g m–2. However, it took 5 d longer to reach this weight than when nutrient supply was at the optimum (Fig. 7B). The effects of other reductions in nutrient supply on the root : shoot ratio and duration of growth required to meet a given shoot dry weight were similarly estimated (Fig. 7).


Figure 2407
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 7. (A) Calculated dependence of Rsb/Ro on time for pansy seedlings grown with different proportions (0·2, 0·4, 0·6, 0·8) of the optimum nutrient supply. (B) The corresponding relationships between Ws and time. Calculations are with the mean temperatures and Par values for the March04, April05 and June05 experiments.

 
Short-cut method for optimizing nutrient concentrations in the irrigation solution
The optimum nutrient concentration in the irrigation solution depends both on the nutrient demand of the seedlings and on evapotranspiration. Increased evapotranspiration increases the volume of nutrient solution supplied (Fig. 1), and thus the quantities of nutrients available for plant uptake. On the other hand, increased evapotranspiration is also associated with increased plant growth, and thus increased plant demand for nutrients. It seemed therefore that these opposite effects could counterbalance one another in some growing systems, so that it may be possible to assess nutrient concentrations without having to make more detailed measurements of evapotranspiration rates. This possibility was tested for each of the four experiments. Daily increments in shoot dry weights were calculated from a starting shoot dry weight by substituting the average value of K2 (g m–2 d–1), for the given experiment in eqn (1), repeating the calculation for each day, and updating Ws accordingly. The corresponding optimum %N was calculated for each day by substituting the value of Ws in eqn (3) and crop N demand was obtained by substituting both Ws and optimum %N in eqn (4). Division of N demand by evapotranspiration for the given day (by analogy with eqn 5) gave the N-concentration required in the irrigation water to meet crop N-demand for that day.

Graphs of N-concentration against Ws prepared in this way for the Mar04, April05 and July05 experiments, using smoothed evapotranspiration data (Fig. 8), were almost coincident until the shoot dry weight exceeded about 50 g m–2. Higher concentrations of N were required throughout growth for the Jan05 experiments because transpiration frequently fell to low values. For the Mar04 and April05 experiments, graphs of N-concentration against Ws were coincident throughout the entire range. It appears therefore that for some growth conditions, N concentrations in the irrigation solution could be adjusted during much of the growth period to meet crop demand by measuring Ws alone at intervals, and omitting the detailed transpiration measurements required for the procedure described in this paper. This would reduce the costs of estimating N concentrations required for optimum growth.


Figure 2408
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		FIG. 8. Relationships between the calculated optimum N concentrations required in the irrigation solution and shoot dry weight during growth of (A) pansy and (B) petunia in each of the four experiments.

 


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
APPENDIX 1
 APPENDIX 2
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 
A method was developed for adjusting nutrient supply of bedding plants according to the changing crop demand in contrast to maintaining the same concentrations throughout growth. It depended on measuring Ws at the beginning and end of each week and the average daily evapotranspiration over the week from which estimates were made of nutrient requirements for the following week.

The new method of nutrient supply enabled growth to be maintained for long periods at different values of the growth rate coefficient [K2 (g m–2 d–1)] without the appearance of any nutrient deficiency symptoms even when nutrients were severely limiting growth. The value of K2 (g m–2 d–1) for the optimum nutrient supply was well defined by a modification of a previously published equation in terms of average temperature and incoming radiation. The ratio of K2 (g m–2 d–1) for a sub-optimum rate of supply relative to that with the optimum rate was logarithmically related to the nutrient supply as a fraction of the optimum.

When the aerial environment was not limiting growth, the root : shoot ratio for the sub-optimum nutrient supply divided by that for the optimum nutrient supply was proportional to the ratio of shoot dry weight for the optimum supply relative to that for the sub-optimum supply.

A simulation model derived from the forgoing relationships enabled the effects of nutrient supply, mean temperature and incoming radiation on growth, and the root : shoot ratio to be calculated for both pansy and petunia.

A further simplification of the new procedure, for some growing situations, enables the concentrations of nutrients in the irrigation solution to be estimated from measurements of Ws alone, without measurements of evapotranspiration, at intervals during growth.

A well-tested algorithm for predicting optimum nutrient concentrations during growth of pansies and petunia has been developed and can be downloaded from the Internet at: http://www2.warwick.ac.uk/fac/sci/hi2/research/plantmineralnutrition/bedmod.


   APPENDIX 1
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 APPENDIX 1
APPENDIX 2
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 


View this table:
[in this window]
[in a new window]

  		TABLE A1. Definition of symbols with default values

 


   APPENDIX 2
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 APPENDIX 1
 APPENDIX 2
ACKNOWLEDGEMENTS
 LITERATURE CITED
 


View this table:
[in this window]
[in a new window]

  		TABLE A2. Definition of symbols without default values

 


   ACKNOWLEDGEMENTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 APPENDIX 1
 APPENDIX 2
 ACKNOWLEDGEMENTS
LITERATURE CITED
 
We thank the UK Department of Environment and Rural Affairs for supporting this work with grant HH1413 SMU, Findons Fine Plants Limited, of Stratford-upon-Avon for sowing and germinating the seedlings, Bulrush Peat for supplying the peat and carrying out chemical analyses, Mrs Marion Eavers for help in this work and for advice, Dr Allen Langton for providing data, and Mr Neil Bragg and Mr Stuart Coutts for support and advice.


   FOOTNOTES
 
{dagger} Present address: University of Plymouth, Faculty Colleges, Duchy College, Camborne, Cornwall TR14 0AB, UK Back


   LITERATURE CITED
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 APPENDIX 1
 APPENDIX 2
 ACKNOWLEDGEMENTS
 LITERATURE CITED
 

    Ågren GI and Franklin O. (2003) Root:shoot ratios, optimization and nitrogen productivity. Annals of Botany 92:795–800.[Abstract/Free Full Text]

    Belanger G and Richards JE. (1999) Relationships between P and N concentrations in timothy. Canadian Journal of Plant Science 79:65–70.[Web of Science]

    Brewster JL and Sutherland RA. (1993) The rapid determination in controlled environments of parameters for predicting seedling growth rates in natural conditions. Annals of Applied Biology 122:123–133.[CrossRef][Web of Science]

    Broadley MR, Bowen HC, Cotterill HL, Hammond JP, Meacham MC, Mead A, et al. (2004) Philogenetic variation in the shoot mineral concentration of angiosperms. Journal of Experimental Botany 55:321–336.[Abstract/Free Full Text]

    Brouwer R. (1962) Nutritive influences on the distribution of dry matter in the plant. Netherlands Journal of Agricultural Science 10:399–408.[Medline]

    Ericsson T. (1995) Growth and shoot:ratio of seedlings in relation to nutrient availability. Plant and Soil 169:205–214.[CrossRef]

    Ericsson T and Ingestad T. (1988) Nutrition and growth of birch seedlings at varied relative phosphate addition rates. Physiologia Plantarum 72:227–235.[CrossRef]

    Greenwood DJ and Draycott A. (1989) Quantitative relationships for growth and N content of different vegetable crops grown with and without ample fertilizer-N on the same soil. Fertilizer Research 18:175–188.[Web of Science]

    Greenwood DJ and Stone DA. (1998) Prediction and measurement of the decline in the critical-K, the maximum-K and total cation plant concentrations during the growth of field vegetable crops. Annals of Botany 82:871–881.[Abstract/Free Full Text]

    Greenwood DJ, Cleaver TJ, Loquens SMH, Niendorf KB. (1977) Relationship between plant weight and growing period for vegetable crops in the United Kingdom. Annals of Botany 41:987–997.[Abstract/Free Full Text]

    Greenwood DJ, Cleaver TJ, Turner MK, Hunt J, Niendorf KB, Loquens SMH. (1980a) Comparison of the effects of potassium fertilizer on the yield, potassium content and quality of 22 different vegetable and agricultural crops. Journal of Agricultural Science, Cambridge 95:441–456.

    Greenwood DJ, Cleaver TJ, Turner MK, Hunt J, Niendorf KB, Loquens SMH. (1980b) Comparison of the effects of nitrogen fertilizer on the yield, nitrogen content and quality of 21 different vegetable and agricultural crops. Journal of Agricultural Science, Cambridge 95:471–485.

    van Iersel M. (1999) Fertilizer concentration affects growth and nutrient concentration of subirrigated pansies. HortScience 34:660–663.[Abstract/Free Full Text]

    van Iersel MW, Beverley RB, Thomas PA, Latimer JG, Mills HA. (1999) Fertilizer effects on the growth of Impatiens, Petunia, Salvia, and Vinca plug seedlings Bedding Plant Foundation Research Report F-9904.

    Ingestad T and Lund AB. (1979) Nitrogen stress in birch seedlings. 1. Growth technique and growth. Physiologia Plantarum 45:137–148.[CrossRef]

    James EC and van Iersel MW. (2001) Fertilizer concentration affects growth and flowering of sub irrigation petunias and begonias. HortScience 36:40–44.[Medline]

    Kang J-G and van Iersel MW. (2001) Interactions between temperature and fertilizer concentration affect growth of subirrigated petunias. Journal of Plant Nutrition 24:753–765.[CrossRef][Web of Science]

    Knecht MR and Göransson A. (2004) Terrestrial plants require nutrients in similar proportions. Tree Physiology 24:447–460.[Abstract]

    In Lemaire G (Ed.). Diagnosis of the nitrogen status of crops (1997) (Springer, Berlin).

    Scaife A, Cox EF, Morris GEL. (1987) The relationship between shoot weight, plant density and time during the propagation of four vegetable species. Annals of Botany 59:325–334.[Abstract/Free Full Text]

    Scoggins HL, Bailey DA, Nelson PV. (2002) Efficacy of the press extraction method for bedding plant plug nutrient monitoring. HortScience 37:108–112.[Web of Science]

    Smolders E and Merckx R. (1992) Growth and root–shoot partitioning of spinach plants as affected by nitrogen supply. Plant, Cell and Environment 15:795–807.[CrossRef]

    Smolders E, Buysse J, Merckx R. (1993) Growth analysis of soil-grown spinach plants at different N-regimes. Plant and Soil 154:73–80.[CrossRef][Web of Science]

    Stamps RH. (2000) Management of nutrients in ornamental plant production systems in Florida: an overview. Soil Science and Crop Science Society of Florida Proceedings 59:27–31.

    Sterner RW and Elser JJ. (2002) Ecological stoichiometry: the biology of elements from molecules to the biosphere(Princeton University Press, Princeton, USA).

    Thornley JHM. (1972) Balanced quantitative model for root:shoot ratios in vegetative plants. Annals of Botany 36:431–441.[Abstract/Free Full Text]

    Tinker PB and Nye PH. (2000) Solute movement in the rhizosphere(Oxford University Press, Oxford).

    Von Arnim A and Deng XW. (1996) Light control of seeding development. Annual Review of Plant Physiology and Molecular Biology 47:215–243.[CrossRef][Web of Science][Medline]

    Zandstra JW and Liptay A. (1999) Nutritional effects on transplant root and shoot growth – a review. Acta Horticulturae 504:23–31.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Greenwood, D. J.
Right arrow Articles by Mulholland, B. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Greenwood, D. J.
Right arrow Articles by Mulholland, B. J.
Agricola
Right arrow Articles by Greenwood, D. J.
Right arrow Articles by Mulholland, B. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?